乌鸦传媒

Patients with newly diagnosed type 2 diabetes had similar improvement in glycemic control whether treated with acarbose (Precose) or metformin, investigators in a randomized trial reported.

After 48 weeks of treatment, the acarbose group had a mean reduction in glycated hemoglobin (HbA1c) of 1.11% compared with 1.12% for the metformin groups. A between-group difference of 0.02% was observed after 24 weeks of follow-up.

Patients in the two treatment groups had identical rates of hypoglycemia and serious adverse events, Wenying Yang, MD, of China-Japan Friendship Hospital in Beijing, and co-authors reported online in The Lancet Diabetes-Endocrinology.

"This study is the first head-to-head comparison of metformin and acarbose as initial therapy for type 2 diabetes after failure of therapeutic lifestyle modification," the authors concluded. "We report that both acarbose and metformin are well tolerated and have similar efficacy as initial therapy for HbA1c reduction in Chinese patients with type 2 diabetes."

"Metformin should remain as first-line treatment for patients with newly diagnosed type 2 diabetes, while patients with exaggerated postprandial excursion can be treated with an alpha-glucosidase inhibitor as an alternative therapy before cardiovascular benefits of acarbose are validated and confirmed in ongoing studies," they added.

The alpha-glucosidase inhibitor acarbose has approval for treatment of type 2 diabetes as an adjunct to diet and lifestyle management. Both acarbose and metformin are widely used as treatment for type 2 diabetes, and acarbose, in particular, is used extensively in China and other Asian countries, the authors noted in their introduction. In contrast, metformin has been used primarily in white populations, and the results have been extrapolated to other populations.

To date no clinical trial directly comparing acarbose and metformin has been reported, although a trial in patients with impaired glucose tolerance showed similar efficacy for the two drugs in preventing progression to type 2 diabetes. Investigators in another study showed similar effects of acarbose and metformin on HbA1c in patients with more advanced diabetes, the authors continued.

Yang and colleagues performed a randomized noninferiority trial to examine the relative efficacy of acarbose and metformin in Chinese patients with newly diagnosed type 2 diabetes. Investigators at 11 sites in China enrolled patients ages 30 to 70, who met World Health Organization criteria for type 2 diabetes and who had not been treated with medication or had been treated for no more than a month.

Eligible patients had an HbA1c of 7% to 10%, fasting plasma glucose ≤200 mg/dL, and a body mass index of 19 to 30.

The primary endpoint was change in HbA1c from baseline to 48 weeks.

The final analysis included 784 patients. The trial began with 788 patients, four of whom did not receive therapy. During follow-up, 16% of patients in the acarbose group and 20% in the metformin group discontinued treatment. After 24 weeks investigators had the discretion to add insulin secretagogues for patients with inadequate glycemic control; five patients in the acarbose arm and four in the metformin arm received the add-on therapy.

An interim analysis after 24 weeks of follow-up showed mean reductions in HbA1c values of 1.17% in the acarbose arm and 1.19% in the metformin group, declining to a difference of 0.01% after 48 weeks. The trial had an upper limit for noninferiority of 0.3% and a lower limit of -0.1%.

The magnitude of HbA1c reduction differed significantly within treatment groups according to baseline HbA1c (P<0.0001) but not between the two groups. A similar proportion of patients in each group achieved HbA1c ≤6.5% at 24 weeks (67% versus 69%) and at 48 weeks (62% versus 64%). Additionally, results did not differ by body weight, BMI, or percentage of carbohydrate in the diet.

A standard meal test did show a significant difference favoring acarbose in area under the curve for serum insulin at 180 minutes at both 24 weeks (P=0.0001) and 48 weeks (P=0.0047).

The authors acknowledged several limitations of the trial: absence of a placebo control group, no use of hyperglycemic clamp or duodenal nutrient perfusion to eliminate differences in blood glucose and gastric emptying, and no assessment of various gut hormones that might be affected by the two therapies.

The author of an accompanying commentary said the trial supports use of acarbose as an alternative to metformin as first-line therapy. However, he joined the authors in noting the need for longer follow-up in patients treated with acarbose or with some of the newer hypoglycemic agents.

"There are currently little medium-term data for the other agents being compared with metformin as monotherapy," said Ronald C.W. Ma, MD, of Prince of Wales Hospital in Hong Kong. "Additional data for the medium-term and long-term outcome of patients receiving these newer agents compared with metformin are eagerly awaited."

"[The trial] represents an applaudable contribution and important step towards the development of evidence-based and population-specific treatment guidelines for type 2 diabetes," he added.

Comment